Reconceptualizing Interorganizational Collaborations

Reconceptualizing Interorganizational Collaborations as Tensile Structures: Implications of Conveners’ Proactive Tension Management

Accepted for Publication in Communication Monographs

DaJung Woo School of Communication Studies University of Tennessee, Knoxville (Corresponding Author: [email protected])

Acknowledgement This work was supported by the National Science Foundation (grant number SES-1057148).

Citation: Woo, D. (2018). Reconceptualizing interorganizational collaborations as tensile structures: Implications of conveners’ proactive tension management. Communication Monographs. Advance online publication. doi:10.1080/03637751.2018.1526389

1 Abstract Conveners, as the main organizers of complex inter-organizational collaborations (IOCs), experience tensions as they make decisions based on collaborators’ competing interests and ideas. This paper theorizes conveners’ tension management as their proactive efforts to shape the IOC processes – as opposed to reactive responses to emergent tensions – and examines how they are related to IOCs’ collaborative capacity. A comparative case analysis of two IOCs in regional planning reveals that conveners’ organizing practices that actively promoted tensions contributed to creating a more dynamic and tension-resistant collaborative environment, compared to those of conveners who tried to prevent tensions. Using tensile structure as a metaphor, the author theorizes about when and how proactively promoting tensions can enhance collaborative capacity. Keywords: inter-organizational collaboration; conveners; tension management; organizational communication; organizational tensions; collaborative capacity

2 Reconceptualizing Interorganizational Collaborations as Tensile Structures: Implications of Conveners’ Proactive Tension Management As organizations face adaptive challenges in today’s competitive and fast-changing environment, they are often advised to explore new opportunities while exploiting existing capabilities at the same time (O’Reilly & Tushman, 2013). Yet, hosting multiple, even contradictory structures and processes within an organization can make their members feel as though they are pulled in inconsistent directions. The clash of ideas or actions—and their interplay—that the members likely experience is referred to as tensions (Fairhurst, Cooren, & Cahill, 2002). Even though the popular use of the term often highlights discomfort or mental struggles that can be caused by tensions, organizational scholars have offered compelling arguments that tensions are routine building blocks of organizational life that are not necessarily negative (Trethewey & Ashcraft, 2004; Gibbs, 2009; Tracy, 2004). Research on organizational tension management over the last decade has consistently shown that embracing tensions, instead of removing them, is critical for success of organizations and their members (see Putnam, Fairhurst, & Banghart, 2016 for a review). However, it is unclear if and how tensions may be proactively set up by organizations that wish to benefit from the push-pull dynamics, as compared to managing emergent tensions reactively. There are two main reasons for this limitation. First, past research has focused on communicative reactions to tensions from the perspective of individuals who have little control or say in organizational decisions, such as how employees construct their own selfidentities in light of competing employer expectations (Pepper & Larson, 2006) and how they react to contradictory interpretations of a newly implemented technology (Jian, 2007). Second, such studies tend to identify organizational practices, such as a merger and a new technology implementation, to the extent that they trigger or uncover tensions to which employees must respond. Consequently, tensions are often painted as unintended consequences of organizational practices, and tension management as individuals’ reactive responses to the resulting situations that impact them.

3 However, examining tension management from the perspective of those who are in charge of organizing can reveal how they might set up tensions proactively to achieve their goals. In fact, studies have shown why organizers might purposely move in competing directions to promote push-pull dynamics. For example, Koschmann and Laster (2011) showed how a neighborhood association included young, White professionals as its members, even though older members viewed them as a threat and the main cause of gentrification. Their meetings involved a lot of disagreements, but the young and older residents’ divergent perspectives motivated and sustained active member participation. Canales (2013) detailed how organizations that allowed both standardization and flexibility in employees’ enactment of organizational rules performed better than companies where rules were strictly enforced, because working through the tensions generated from diverse rule enactment behaviors allowed them to reflect on their broader organizational goals and decisions. As these examples illustrate, organizers can incorporate tensions intentionally to enhance their goal pursuits, rather than relying on reactive strategies when/if tensions arise. Although promoting tensions and their push-pull dynamics can lead to productive outcomes, the costs and complexities associated with maintaining tensions can make the organizers believe that preventing tensions is a more rational and effective way to prepare for their organizational processes. The aforementioned study by Canales (2013), for example, admittedly described how the organization that successfully promoted the standardization--flexibility tension had longer meetings and involved more contentions, making them more difficult to manage. Especially when organizers have time pressures or pressures to produce certain outcomes, they may prefer efficiency over exploring tensions (Dean & Oetzel, 2014) and try to remove the potential sources of tensions. In other cases, social and professional norms can motivate individuals to avoid diverging from the typical ways in which things are done (Apker, Propp, & Ford, 2005) by privileging one option over the other. For these reasons, organizers can rightfully work to prevent tensions that can arise and complicate their processes.

4 The goal of this paper is to examine and compare organizing practices aimed at preventing versus promoting tensions, and their impacts on interorganizational collaborations (IOCs)—processes in which conveners, as the main organizers, bring together relevant organizations and gather inputs to move toward their collective goals (Wood & Gray, 1991). Through a comparative case analysis of conveners’ proactive tension management in two IOCs organized for urban planning, the study reported herein makes three major contributions. First, it offers an alternative perspective to rather passive tension management strategies that have been suggested (e.g., accept and live with tensions; Lewis, Isbell, & Koschmann, 2010) and demonstrates how tensions can be encouraged (or discouraged) proactively through organizing practices. Second, the paper offers a novel theoretical framework called tensile structure, which compares organizing IOCs to building a construction that uses strengths generated from tensions. This reconceptualization will offer ways to theorize visually how promoting tensions, rather than preventing them, can enhance collaborative capacity of IOCs. Third, the study provides practical implications for conveners to be mindful about the dilemmatic nature of their role and how their practices might shape IOCs in relation to tensions. Theoretical Background Tensions in Interorganizational Collaborations Organizational processes within a single organization often involve tensions as its members exchange and attempt to make sense of incongruent messages about various issues (Fairhurst et al., 2002). It is not surprising, then, that collaboration among multiple organizations with different practices and values involve tensions that are complex and consequential (Gray & Purdy, 2014). Past research has identified tensions as a key issue in various IOC contexts, including community health (Medved et al., 2001) and refugee networks (Tomlinson, 2005); however, most tensions discussed in the IOC literature are those that collaborators face as they engage in joint activities, such as making decisions together while having to compete for limited resources (Doerfel & Taylor, 2004). Although those tensions reveal

5 the complexities of communication and relationships among collaborators, the role of conveners in managing tensions has been largely overlooked. Empirical studies on conveners are limited, but scholars have recognized that their practices shape IOCs and collaborative outcomes in important ways (Dorado & Vaz, 2003; Purdy, 2012). As the organizers of IOCs, whose work impacts and is impacted by multiple entities, conveners would be no exception to the experience of tensions. However, conveners are likely to manage different types of tensions compared to those that collaborators experience. Consider how supervisors are positioned differently to deal with broader managerial and structural issues due to their organizing power and responsibility (e.g., tensions arising from new executive mandates and subordinates’ expectations during restructuring; Lüscher & Lewis, 2008) compared to their employees. Similarly, as conveners organize IOCs, they are likely to deal with structural tensions related to membership boundaries, rules of operation and engagement, and resource distribution (Lewis et al., 2010). Structural tensions that conveners have to manage are different than tensions arising from collaborators’ interactions and may or may not be visible to collaborators— although they can be related—because of their convening power to design the process (Gray, 1989) which collaborators do not have. Given the lack of empirical evidence on conveners’ work and the unique types of tensions they manage as the organizers of IOCs, the first research question explores: RQ1: What are the major types of tensions that conveners manage when organizing for IOCs? Tension Management Strategies As communication scholars have long argued, people’s experiences are not shaped by the existence of tensions per se, but by how they manage and respond to tensions communicatively (Baxter & Montgomery, 1996; Driskill, Meyer, & Mirivel, 2012; Tracy, 2004). Accordingly, it is important to examine conveners’ tension management practices to understand the implications of tensions on IOCs. A recent metatheoretical work by Putnam et al. (2016) offers insights into three major approaches that conveners might use for managing tensions as they organize for IOCs. Since most previous research

6 reviewed by Putnam et al. suggests reactive strategies, I re-group the three approaches into two categories based on the motivations for using the approaches proactively: (1) to prevent tensions (eitheror approaches) and (2) to promote tensions (both-and and more-than approaches). First, if conveners are motivated to prevent tensions, they are likely to take either-or approaches (Putnam et a., 2016). Either-or approaches treat oppositional tension poles as independent and deny their co-existence; thus, people with either-or orientation tend to separate tensions and favor one over the other (i.e., selection). This can be an appealing option to conveners because it offers the potential to prevent challenges that tensions can cause and allows them to organize the step-by-step process of IOCs neatly (Medved et al., 2001). For example, to manage formalization-flexibility tension, conveners may choose formalization to develop a clear structure, set a boundary, and increase efficiency; later, they may realize that the IOC process takes many directions and that their abilities to adapt to unexpected situations are limited. In such a situation, using an either-or approach can invoke defensive reactions, such as denying the need for flexibility and disengaging, which can cause long-term dysfunction (Smith & Lewis, 2011). Thus, conveners who prevent tensions through either-or strategies are likely to shape IOCs that limit newly emerging opportunities (Tracy, 2004). Second, if conveners are motivated to promote tensions, they are likely to engage in both-and or more-than strategies (Putnam et al., 2016). Both-and approaches, such as vacillating between tension poles or balancing them, treat tensions as interdependent. Although research has shown that both-and practices are more productive than either-or approaches, they are difficult to implement successfully and often work temporarily (Putnam, 2015). For example, even if conveners decide to balance formalizationflexibility tension, collaborators may develop preference for one over the other, or conveners may fail to keep the tensions balanced throughout the IOC. Then, despite conveners’ intention to utilize a both-and approach, the IOC process can eventually turn into an either-or situation. Further, both-and approaches rely on oppositional tensions as available options without exploring other possibilities. Thus, promoting

7 tensions through both-and practices can lead to IOCs that may not be able to maximize synergy from the interplay of tensions to its full potential. Another way to promote tensions is through more-than practices (Putnam et al., 2016). As the label suggests, this approach moves beyond the opposite tensions (i.e., transcendence) by situating opposite poles in new ways such that they are not pitted against one another (i.e., reframing), or creating a third option or a third space where people can explore tensions discursively and transform ideas (Christensen, Morsing, & Thyssen, 2015). For example, to manage identity tensions (we- versus meorientation; Lewis et al., 2010), conveners can promote creating a third identity that focuses on neither of the competing options (see Woo, Putnam, & Riforgiate, 2017). Conveners who engage in these practices are likely to encourage collaborators to reframe tensions as opportunities and navigate them to generate new options, so they can maximize the collaborative potential of the IOC. However, more-than practices require time and skills to direct the tensions toward collaborative goals. To investigate whether and how conveners try to prevent or promote tensions, examining their organizing practices will be necessary. Researchers who take the tension perspective can identify participants’ different approaches to tensions not because people report their explicit recognition of tensions and intentional (re)actions to them, but based on how they describe their routine activities and reasons for their actions (see Tracy, 2004). Thus, conveners’ proactive tension management will be reflected in the ways in which they describe and rationalize the decisions they make when organizing for their IOCs, such as why they bring certain organizations to the table or why they distribute certain resources, and not others. Taken together, the second and third research questions ask: RQ2: Through which organizing practices do conveners manage tensions identified in RQ1? RQ3: How do conveners engage in the organizing practices (identified in RQ2) differently when they try to prevent versus promote tensions? Interorganizational Collaborations as Tensile Structures

8 Whether conveners try to prevent or promote tensions will, in turn, influence how collaborators work toward their shared goals. Although scholars have broadly recognized tensions as potential sources of productivity for enabling and sustaining collaborative actions (e.g., Koschmann & Laster, 2011) and thus suggested that promoting tensions would be important, the extant literature lacks a concrete framework to theorize the mechanism through which tensions can support IOCs. To fill this gap, I use tensile structure as a metaphor to theorize why promoting tensions helps to maximize IOCs’ collaborative potential. Tensile structures are a type of construction for making tent-shaped ceilings, which requires tensions generated from pulling the used materials into different directions (see Figure 1). That is, one cannot build a tensile structure without tensions that support its overall shape. Below, I discuss three main components of this metaphor to conceptualize IOCs as tensile structures. [Figure 1 about here] The first key component of tensile structures is the materials used. Building tensile structures requires materials that have high tensile strength (i.e., the amount of tension a material can withstand without breaking while it is pulled into opposite directions). Using aluminum, for example, will not be successful because it will easily break under tensions, compared to materials like fabric. The material can be compared to the main topic that an IOC is intended to address. IOCs are often organized to tackle a focal issue with varying levels of specificity. An IOC organized to address a “wicked problem” (Weber & Khademian, 2008) that does not have definite causes or solutions (e.g., improving public health) has high tensile strength because it allows for deliberating about the main concern from diverse perspectives—or “pulling” it into different directions, figuratively speaking—compared to an IOC organized to accomplish a particular outcome (e.g., creating a specific health report). If conveners promote tensions when an IOC has low tensile strength, tensions are not likely to serve the purpose of the IOC well but can contribute to its “breaking” under strain; but, promoting tensions in IOCs with high tensile strength can help to achieve the collaborative goals through generating creative ideas.

9 The second and third components of the metaphor are the forces that pull the material into different directions (the hands in Figure 1) as conveners’ tension management, and the resulting shape of a tensile structure as collaborative capacity1 of an IOC. If conveners try to prevent tensions through either-or approaches, they would only select one of the opposite poles to support IOCs; as a result, the tensile structure will have a collapsed corner due to the loss of tension, decreasing the space that the structure can create (see Figure 2). The limited space signifies a reduced discursive space where collaborators can engage with tensions. By contrast, if conveners promote tensions, they would attend to or “pull” both of the opposite poles. In doing so, using more-than approaches will help to maximize the space better than both-and approaches because both-and approaches, such as vacillating, are not stable in keeping the tension poles in place (e.g., vacillating involves focusing on one pole while the other pole is “loosened,” causing temporary collapses of a tensile structure). More-than approaches recognize that, by pulling both tension poles simultaneously, IOCs as tensile structures can create space in which collaborators can keep the tensions open and seek energy from them to potentially move beyond the available options, while valuing both poles equally (Putnam, 2015). [Figure 2 about here] Taken together, I posit that, when the collaborative goal is to address wicked problems (high tensile strength), conveners who promote tensions through more-than approaches will organize IOCs with higher collaborative capacity than conveners who prevent tensions through either-or approaches. In the tensile structure metaphor, conveners’ tension management is necessarily proactive because the space (the analogy for collaborative capacity) cannot be created until tensions are set up; thus, conveners’ tension management is conceptualized to make a difference in IOCs’ collaborative capacity. To explore these ideas inductively, the fourth research question asks:

As the term has been defined inconsistently across the literatures, I follow traditional uses of the term “capacity” and refer to the ability for an IOC to generate diverse ideas or resources toward achieving its collaborative goal. 1

10 RQ4: How do conveners’ different tension management practices (preventing versus promoting tensions) shape IOCs with different levels of collaborative capacity? Methods Research Background This study draws from my year-long field research on regional planning in two major metropolitan areas in the United States (“Region A” and “Region B”). Regional planning is an ideal context for this study because it relies on IOCs facilitated by designated conveners, who are urban planners at Metropolitan Planning Organizations (MPO). Conveners organize IOCs for a variety of purposes, but one of the important reasons is to facilitate the development and refinement of their region’s long-range plans. Long-range plans describe visions and future directions for the region’s developments in the next 25-30 years, and they are updated by urban planners every four years based on the inputs gathered from various stakeholders. While the urban planners, as conveners, have to meet certain mandates (e.g., how many meetings to hold), it is within their discretion how to organize their IOC processes. The participating organizations (collaborators) share the same broad goal of improving the region’s urban environment; yet, they have different expectations for what a ‘good community’ should look like. Thus, the IOC processes involve tensions around how collaborators negotiate their different ideas and perspectives. The fact that regional planning requires diverse inputs and constructive criticism characterizes the IOCs as having high tensile strength—meaning that the IOCs can not only endure but also need some level of tension to support the development of good regional plans. Conducting the study in two different sites allowed for comparing the conveners’ practices. The two regions shared similarities (e.g., the fact that their IOCs followed the same federal guidelines, timeline, and used the same collaboration tools), which made them comparable cases. They also shared differences in terms of education levels, key regional issues, and diversity, which indicated that conveners in the two regions might have different approaches to tensions as they organize their IOCs. Data Collection

11 I conducted in-depth interviews with 14 and 23 conveners in Region A and B, respectively (90 percent of the planning staff at each region’s MPO). I used a semi-structured interview protocol with three broad questions about their routine practices for organizing IOCs, goals, and challenges. Since I entered the IOCs during the last year of the four-year cycle, I used retrospective interview questions about how they developed their processes before the cycle began (no interview questions asked directly about tensions). Conveners’ responses to these questions helped to identify their proactive practices based on their anticipated tensions. However, proactive tension management does not mean that it only happens before IOCs begin; as new events and issues emerge in such long-term IOCs, conveners adjust their expectations about how things will happen in the upcoming phase of their IOCs and engage in different practices. Thus, I asked follow-up questions about whether and how their organizing practices shifted over time. Conveners’ reports about why they made certain decisions when multiple options were available (e.g., who to involve) and how they managed situations with potentially conflicting outcomes provided especially valuable data about their proactive approaches to tensions. In addition to conveners, I interviewed collaborators from the two IOCs (38 representatives from 14 different organizations in Region A; 35 representatives from 24 organizations in Region B). I recruited them through snowball sampling until no new organization was mentioned by interviewees. I used a semi-structured interview protocol, which included three major questions about their (a) motivation for participating; (b) relationships and interactions with conveners; and (c) overall thoughts about the IOC process. These interviews allowed me to understand conveners’ practices and their influence on IOCs from the collaborators’ perspectives. As will be discussed later, collaborators’ interview data were used mainly to examine RQ4 and for corroboration purposes when exploring RQ2 and RQ3. All 110 interviews were recorded and transcribed verbatim. During the field research, I also conducted nonparticipant observations at various IOC meetings and workshops for approximately 20 hours in total, where I took field notes based on the observations and my critical reflections.

12 Data Analysis Tensions in conveners’ organizing practices became salient about two thirds of the way through my field research, which then became the focal point of my data analysis. Below, I summarize how the interview data were analyzed to explore the four research questions. Even though I present the analytic processes as separate stages for clarity, they occurred iteratively and informed one another. Identifying types of tensions managed by conveners (RQ1). To explore RQ1, I analyzed 37 conveners’ interviews to identify what types of tensions are related to their organizing practices across the two IOCs. Specifically, I focused on their responses to the questions regarding routine decisions they make as conveners, as well as challenges associated with them. When reading each transcription, I coded for convening activities that involved competing options and expectations as described by interviewees. For example, when conveners explained their anticipation that some collaborators might be unhappy to see certain items on the meeting agenda while others would expect to discuss them, it was coded as “preparing for IOC meetings.” This process involved creating analytic memos with ideas about how the codes were related to one another, allowing for open and axial coding to occur in tandem (Strauss & Corbin, 1998). As my analytic memos generated new themes, I engaged in constant comparison with the data to refine and confirm the themes. As a result, I identified three major categories of convening practices that involved tensions: (a) deciding who should participate; (b) determining the focus of IOCs; and (c) arranging meetings with collaborators. Specific types of tensions under each category will be discussed in the findings. Conveners’ tension management practices (RQ2, RQ3). After I identified the three categories of tensions, I went back to the 37 conveners’ interviews to examine which specific organizing practices facilitated their management of those tensions (RQ2). Using the open codes from the previous stage as guides, I analyzed how conveners rationalized key decisions they made as they tried to manage situations involving the three types of tensions; for example, when they expected some collaborators’

13 dissatisfaction with certain items on the agenda (tension related to “determining the focus of IOC”), did they include or remove the items, and why? Analyzing the justifications of conveners’ such decisions helped to tap into their proactive tension management, as they revealed how the conveners intended to organize IOCs in anticipation of tensions that would likely influence future meetings and interactions. After generating codes related to various reasons for conveners’ decisions aimed at managing the three types of tensions, I consulted relevant literatures to connect the emergent codes with knowledge drawn from existing research. This process resulted in three major practices that facilitated conveners’ management of the three tensions identified in RQ1 respectively: (1) identity work; (2) anticipatory work; and (3) boundary work. Until this point, I have not considered differences between Region A and B since RQ1 and RQ2 were related to common tensions and practices across the two IOCs. To explore RQ3, however, I treated conveners from Region A and B as two separate cases to examine how they engaged in the three organizing practices differently to manage the identified tensions. I conducted within-case analyses (Yin, 2014) by reading conveners’ transcriptions from Region A and B separately, while noting how their use of the three practices reflected their different orientations toward the tensions. This involved iterative processes of going back and forth between my data, analytic memos and field notes. It was during this process that my hunches from the field research about the conveners’ different approaches to tensions were confirmed: conveners in Region A tried to prevent tensions, whereas conveners in Region B promoted tensions. Once I identified this pattern, I solicited informant feedback from conveners.2 The conveners reported feeling enlightened by the tensions I presented and agreed that the three organizing practices made differences in IOC processes in relation to tensions.

2

This member check process involved conveners from Region A only. The MPO in Region B was going through major restructuring, and the lead convener who was the key informant for this study had left the organization.

14 Comparison of the two IOC’s collaborative capacity (RQ4). Finally, to compare how conveners’ different tension management practices affected the two IOCs’ collaborative capacity, I used both conveners’ and collaborators’ interview data to examine three factors: (a) collaborator diversity; (b) levels of consensus; and (c) convener readiness. I determined that these were important indicators of the IOCs’ collaborative capacity because the main goal of IOCs in regional planning was to gather as diverse inputs as possible from a broad cross-section of the public so as to best inform their planning activities. Thus, I expected that IOCs that were organized to have high collaborative capacity in this context (i.e., conveners promoted tensions instead of preventing) would have highly diverse groups of collaborators; low levels of consensus due to the expectation for collaborators to offer diverse perspectives and challenge each other’s ideas; and high convener readiness to engage with collaborators (see Table 1 for theoretical and operational definitions). With an exception of coding for collaborator diversity, which involved simple counting of the interviewees without planning- and government-related backgrounds and calculating the percentage, I utilized consensus coding with two research assistants who had not participated in site visits. In a series of extensive discussions of the data, we generated and refined codes through an inductive-deductive process, and then jointly analyzed conveners’ and collaborators’ interviews. Any disagreements about categorization of codes were resolved through discussions. Table 1 details the final coding schemes that were generated from the consensus coding procedure, as well as the coding outcomes. Overall, Region B had higher collaborative capacity than Region A based on the three factors I examined. [Table 1 about here] Findings and Interpretations This section is organized in three parts. First, I define and illustrate three major tensions that conveners managed in both Region A and B. Then, I contrast the two cases to show how the conveners

15 managed the tensions differently through three organizing practices. Lastly, I discuss how the conveners’ different approaches to tensions contributed to the two IOCs’ collaborative capacity. Types of Tensions Managed by Conveners Tension 1: Who should participate? Objective experts versus biased experts. The first tension was related to conveners’ decisions about who should participate in the IOC. Although most conveners stated that anyone from their region could share feedback on regional plans, they also recognized that involving certain groups in the IOC could cause challenges. Specifically, conveners described two types of collaborators: those who shared objective expertise versus biased expertise. The differentiation was not based on the actual expertise of collaborators but based on their affiliation (organizations with/without agendas against existing regional plans) and abilities to discuss technical aspects of planning. Certainly, the two types of experts did not have to be mutually exclusive; but, conveners often interpreted the two types of collaborators’ inputs as incompatible. For example, Jeanine described her frustration when a local activist group questioned her team’s use of quantitative data for producing regional plans and tried to force its agenda: “[We use] state-of-the-art methods… You can’t just sit there and make guesses!” Jeanine was frustrated by how members of the activist group could not appreciate her team’s scientific work due to their lack of technical expertise and their expectation that regional planning involved biased “guesswork.” Yet, conveners like Cara viewed that “objective experts” who relied on data-driven work could lack an understanding of “how things work on the ground,” explaining why their scientific work might seem inconsistent with inputs shared by “biased experts” whose work focused on advocacy through community engagement. Thus, conveners had to manage the objective—biased expert tension when deciding who should join the IOC table. Tension 2: What is the IOC for? Process versus outcome. The second tension was related to determining the focus of the IOC. One of the major responsibilities of urban planners (conveners) was analyzing traffic and socioeconomic data to draw tentative conclusions about what kinds of

16 developments are needed in the future. The results of their analysis then became the basis for IOC activities. Conveners agreed that collaborators’ feedback on the outputs of their data analysis was valuable; however, when collaborators raised questions about the process of data analysis, conveners felt conflicted because explaining their analytic decisions to non-planning experts could cause delays in the IOC, even though they recognized the value of discussing the process with collaborators. For example, Derrick (convener) explained, “When they [collaborators] see a certain output that doesn’t resonate with them, they question, ‘where and how did they [conveners] get this number?’” At the same time, conveners believed that IOC meetings should be organized with average collaborators in mind, rather than those who had the abilities to understand and discuss planning analytics: “An average person is often looking for an answer to a particular question, and what the output means to him or her…summarizing the 40-page document [about the analytic process] would be nice to know, but not what they need to know” (Maya, convener). Thus, conveners had to manage the process—outcome tension to determine what their IOC’s focus should be. Tension 3: How to engage? Formal versus informal. The third tension was related to how conveners engaged with collaborators. Since regional planning involved bureaucratic processes with many regulations (e.g., types of committees and reporting structures), conveners were used to organizing through formal mechanisms to meet the requirements. But, they often found themselves informally engaging with collaborators in unstructured and spontaneous manners. The informal engagements gave rise to a tension because, even though they offered opportunities to exchange additional ideas outside of formal meetings, conveners found them interfering with the time that they could spend doing analytic work and preparing for formal IOC meetings. For example, Chuck said, “I enjoy being able to focus on a [data analysis] project, but I can’t…I feel like I am a chicken with my head cut off because when you’re at the whim of phone calls or questions, it becomes chaotic.” At the same time, conveners wanted to ensure that they remained open to collaborators’ questions or feedback to foster trusting relationships.

17 Anne said, “As we always say, ‘if you [collaborator] want to talk about the data, we are happy to talk about it whenever.” Anne added how conveners might also approach collaborators informally: “If I say something wrong, I might call them [collaborators] the next day and say, ‘you know, when I said this the other day, that was actually wrong.’” Conveners’ management of the formal—informal engagement tension was evident, given their reports that focusing on one could limit the efforts they could make for the other. Region A Conveners’ Tension Management Preventing tension 1: Selecting objective experts through identity work. Conveners in Region A managed the first tension by selecting objective experts over biased experts for their IOC (either-or approach) such that they could prevent tensions that could arise if they involved both types of experts. The main practice that facilitated their tension management was identity work—the daily performance of individuals’ roles and negotiation of interactions to achieve a certain identity (Watson, 2008). Within the conveners’ categorization of experts as objective versus biased, they constructed their own professional identity as objective experts. When considering who to involve in their IOC, the conveners’ discursive positioning of themselves as objective experts guided their decision to select other objective experts as their collaborators. They considered engaging with biased experts a potential threat to their objective expert identity and believed that limiting the involvement of biased experts was a proactive way to organize the IOC effectively. To illustrate, Chrissy (the lead convener) explained why her team did not involve a local community advocacy group when the group came to share its inputs. She emphasized how the group shared “emotional reactions” to regional plans based on their agenda rather than data-based evidence: [Engaging with groups like that] raises a lot of questions if it is an appropriate thing. It's a dangerous place to go; to go from respected unbiased data experts to working with advocates. Advocacy groups don’t want to see the regional plans happen because of their agenda. So, we ask ourselves ‘is it okay to talk to them?’ We want to maintain the reputation of having rigor behind what we put out, rather than emotion—which, everybody’s got emotion, you can't really

18 shut it off. At the same time, we have to be careful with that stuff [whether to engage with them or not]. Chrissy’s comment reflects her team’s collective concern that involving advocacy groups could lower their credibility as objective conveners. Thus, her team had been working with groups known for their technical domain expertise in the region. One of such collaborators was a business development firm. In an interview with the director of the firm (Gary), he said, “I was brutally frank with conveners [when I shared my inputs] because I was telling the truths and they could trust what I was saying.” I could not find evidence in my data that inputs shared by Gary were actually based on objective truths; but, it was the conveners’ interpretation that collaborators like Gary were unbiased and helpful for keeping their identity as objective experts. Chrissy said that working with Gary’s firm “helped to make sure that we [conveners] are in touch with the reality, because it is important for us to know what kinds of business decisions will be made in our region and if what we are planning will be possible.” Region A conveners’ identity as “objective” experts were explicitly noted by 13 collaborators (34 percent). For example, Mike (a local city planner) said, “They [conveners] are data-driven, have an academic focus and an objective image…they are seen as an objective source of information.” Erik (an engineer from the State Department of Transportation) explained that conveners’ and collaborators’ similar levels of technical expertise facilitated discussions in IOC meetings: “Our world is driven by data and statistical modeling to identify trends and make projections…we all get the systems operations side of planning.” Evan (a consultant from a planning firm), who had worked with the conveners for over a decade, shared his insights regarding why the conveners tried to keep their objective expert identity: “They try to produce objective work that can inform policy decisions to be made by elected officials,” as they viewed that it was elected officials’ responsibility to engage with advocacy groups and make decisions about whether to support the groups’ agendas, and not theirs. By not involving biased experts, the conveners tried to prevent their professional identity from being questioned by their stakeholders and to prevent the IOC process from being overcomplicated by divergent agendas.

19 Preventing tension 2: Selecting outcome through anticipatory work. Conveners in Region A managed the second tension by choosing to discuss the outcome of their data analysis over the process (either-or approach) in their IOC. Their proactive selection of outcome was evident in their anticipatory work (Barley, 2015)—the process by which individuals’ anticipation of their future communication with experts from different domains shapes their routine work. When the conveners described how they prepared materials for IOC meetings, which was their routine practice, they revealed their expectation to receive feedback on their data analysis outputs. The main reason for the anticipation was, as noted earlier, the conveners’ belief that the process was too complicated for non-planning experts to comprehend; another reason was that the conveners viewed the analytic process as their professional territory and felt confident about its rigor (as reflected in Jeanine’s quote earlier regarding her use of a “state-of-the-art methodology”). Thus, they did not expect to be questioned about their process. For example, Patrick’s (convener) quote indicates both reasons: It’s rare that people will care about the analytic process, especially when you talk with people who don’t have time or interest to dig into the technical details. They just want to know, ‘What does this mean for me?’ That is why they have us who really know how to do this stuff...so, I use the analysis results as a tool to tell a story about what will happen. As long as we establish that we did it as objectively as possible, people won’t really question the underlying details. The conveners’ anticipation worked as a proactive mechanism for their tension management (selection) by shaping their organizing practices to focus on the outcomes over the process in several ways. First, when the conveners prepared for IOC meetings, their goal was to best summarize the outcome of their analyses and its implications. Patrick said, “We try to build a narrative based on the outputs to say, ‘these are general ways we could try to develop our region’... what we share is a fraction of what we do [to produce the outputs].’” Second, to ensure that the analysis outputs are accepted by collaborators, Chrissy (the lead convener) trained her staff to present the outputs with caution to prevent potential resistance from the collaborators. For example, she told other conveners to say, “Based on our analysis of our region’s challenges and policies, data outputs show that the likely future will look like

20 this,” instead of saying, “We [conveners] want our region to grow in this way.” Third, even though the conveners conducted data analyses individually for different projects, they worked together afterwards to check “if the different parts of the outputs could create a coherent story” (Maya, convener). Consequently, most collaborators had become used to being presented with result summary and believed that discussing the implications of the outputs was more important than knowing the process. For example, Quincy (a local government representative) said, “What’s important is for suburban communities to realize their need to develop… the stories [that conveners tell based on their analysis outputs] become almost like an education tool for [helping collaborators] realize why development is needed.” There were, however, two collaborators who expressed concerns about the conveners’ selection of outcomes. Reid (a local metro center manager) said the conveners’ practice gave them “some power or luxury that we [collaborators] don’t have,” because when they suggested a plan by saying, “Our analysis shows this needs to be done,” it was hard for collaborators to object the plan and propose alternative plans. Vince (a transportation modeler at a consulting firm) argued that the conveners should not think that they “are fulfilling their role by simply sharing the outputs” but needed to do a better job at demonstrating the values and perspectives that led to the outputs. Preventing tension 3: Selecting formal engagements through boundary work. Conveners in Region A managed the third tension by selecting formal engagements for their IOC over informal meetings (either-or approach). Their proactive selection of formal engagements occurred through their boundary work, which refers to “the ways in which people create, maintain, or change boundaries in order to simplify and classify the world around them” (Kreiner, Hollensbe, & Sheep, 2009, p. 705). Like those who manage their work-home boundaries by segmenting the two worlds (e.g., using two separate calendars for work and home), the conveners separated their work space into two dimensions: one where they worked as conveners and the other as data analysts. Then, they organized formal IOC meetings as their primary space to act as conveners so that the IOC did not interfere with their work as data analysts

21 (Chuck’s comment earlier reflected his discomfort related to the burring of the boundaries). Organizing formal meetings made it easier for the conveners to manage the boundary through pre-planning of where/when to meet, who to invite, and what to discuss. Formal IOC meetings were organized around various committees and working groups. Jim (convener), for example, described a citizen’s advisory committee as follows: The committee members had been on it for a long time...things stared to break down over time and the process wasn’t effective for getting inputs from the members...we are not sure what to do about it. We don’t want to just throw the idea [the committee] out. Maybe we can update the membership and call meetings every month or quarter. Even though Jim considered this committee “dying,” his idea of improving the process was through, yet again, convening formal committee meetings but more regularly and with new members. Comments such as Jim’s indicated that formal meetings had been and will likely continue to be the primary space for the IOC in Region A. Interestingly, when asked about the committees they were part of, collaborators described some committees as having informal aspects. For example, Candace (a city planning director) said, “Our committee looks at land-use issues informally, whereas the other committee is a more formal group that makes decisions about projects.” Jane (a city engineer) said, “We informally invite some private consultants to our committee meetings because they can help us learn about the issues.” As Candace and Jane’s quotes indicate, the “informal” discussions still occurred within the boundaries of formal committee meetings that were organized by the conveners. However, the conveners’ proactive selection of formal engagements did not mean that they would deny collaborators’ requests to meet outside of formal meetings. As employees of a government agency, they had the responsibility to listen to their stakeholders’ inputs. But, when collaborators approached informally, the conveners felt uneasy about the spontaneity and considered them interruptions. For example, Kelly expressed her frustration when university researchers from her collaborative network came to discuss her team’s traffic data: “Their ideas were not really well-crafted… after they left, I looked at Chrissy and was like, ‘you’re kidding me, right?’” Although Kelly did meet

22 with the unexpected visitors, she felt reluctant to engage fully in a dialogue with them about their ideas. In another case, conveners tried to re-direct collaborators to formal meetings. When Zain (a community organization member) contacted the conveners to ask questions, the conveners “called us back and said that we can come to their public meeting.” Since Zain wanted to have in-depth discussions with the conveners, he was disappointed about the conveners’ invitation to a public meeting and thought, “They don’t work very hard at it [IOC].” Other than Zain, most collaborators shared neutral reports about attending formal meetings, possibly because they had become used to the format. Region B Conveners’ Tension Management Promoting tension 1: Reframing as “anyone’s right to participate” through identity work. Unlike conveners in Region A who selected objective experts over biased experts for their IOC to maintain their own objective professional identity, conveners in Region B reframed the tension (morethan approach) as “anyone’s right to contribute” to regional planning. Instead of positioning themselves in the objective expert category, the conveners tried to construct their identity more flexibly and realistically as experts who “could make human errors” and thus needed to remain open to others’ feedback to do their job well. This helped them see their role as enabling various organizations to participate in the IOC regardless of their agendas. This approach indicated that the conveners were proactively encouraging the unknown, future tensions that could emerge from conflicting ideas shared by collaborators with vastly different backgrounds. Even though they recognized the challenges of incorporating inconsistent inputs, they did not pit the different expertise against one another but considered them as sources for spotting limitations in the regional plans that they could not see. To illustrate, Tiffany (convener) said that any organizations, “as part of the public, have a very important role to play. It’s their money [tax dollars] and it’s their region... if they want to leave a legacy for the next generation, it’s important that they get involved.” Another convener, Maureen, stated, “[We] try to recognize the contributions of each organization to the table...that’s when the work becomes more

23 productive.” Comments like these reflected the conveners’ belief that the any groups have the right and responsibility to help improve their region by getting involved in the IOC. Further, compared to Region A conveners’ wish to be known for their objectivity, conveners in Region B wanted to be trusted for their openness to criticism. For example, Katie (convener) said: Working with advocacy groups is interesting, because they are adversary in that one of them will very likely sue us, and if they sue us, it gets very complicated. On the other hand, we want them to have the right to sue us because, if we are carrying out duties in a way that is in conflict with the law or harming the environment unnecessarily—or whatever their basis is for the lawsuit— they are entitled to sue us. As Katie’s quote suggests, the conveners believed that involving collaborators who could challenge their work was necessary to improve the regional plans, even if it complicated the process. When collaborators accepted their ideas without questions, the conveners urged them to be more critical. For example, when members of a local nonprofit asked John (lead convener) to clarify how the risks of the regional plans were analyzed, he spent an hour explaining the analytical details. When the director of the nonprofit finally said, “Oh my gosh, that is a lot. Okay, we trust you guys,” John responded by saying, “Don’t just trust us because it’s complicated!” and wanted to engage in a dialogue with them. The conveners’ identity work and promotion of tensions was evident in collaborators’ reports about the IOC dynamics. They discussed how their perspective was different from those of the conveners and other collaborators, but the differences led to constructive changes to the regional plans. For example, Suzy (leader of a community organization) said, “I do think that they [conveners] are smart and have good intentions, but it was appalling that they could not see the impacts of the plan on lowincome communities...but they made some really significant changes based on what came out [of the IOC meetings].” Brian (environmental activist) echoed by saying that working with a variety of groups “helped to shape the regional plan for the better.” Involving organizations like Suzy’s and Brian’s in the IOC would have been seen as “a dangerous place to go” by conveners in Region A.

24 Promoting tension 2: Transcending process and outcome through anticipatory work. Unlike conveners in Region A who focused on the outcomes of their data analyses in IOC discussions, conveners in Region B expected to deliberate about any aspects of the regional plans. They admitted that preparing to cover the process and the outcomes required different kinds of efforts; however, instead of selecting one or the other, they transcended the two (more-than approach) and prepared to show the underlying principles behind their planning work, such that collaborators could have a holistic understanding of the relationship between the process and the outcomes. By doing so, they aimed to empower the collaborators and to encourage them to share informed feedback on any “moving parts” of the regional plans. By not focusing on the process, they could avoid overwhelming the collaborators with technical information; also, by not focusing on summarizing the outputs, they could motivate collaborators to reimagine the regional plans and propose alternatives. My data analysis revealed three ways in which the conveners’ anticipatory work reflected transcendence of the process-outcome tension. First, they worked with the MPO’s internal communication staff to create materials that could promote a high-level understanding of their planning work from a non-planner point of view. The “puzzle pieces” they created visually represented key themes of their planning analyses and how they came together to create the regional plans. Second, the conveners made their data—both raw and processed—sharable, expecting collaborators to want them for reexamining any aspects of their work. When a social justice nonprofit requested the conveners’ data, John (lead planner) let the person “plug on an external hard drive and get all the files so that they could have a deep dive into the data.” Third, the conveners prepared to disclose limitations of their data analyses so that collaborators could interpret the outputs accordingly. For example, Henry explained that, when making presentations for IOC meetings, he prepared to show how “we (conveners) are finalizing a lot of things, and there are a lot of things we still don’t know... we had to assume that certain things are

25 the same everywhere [in our region] because we don’t have that much data.” This contrasted with Region A conveners’ emphasis on demonstrating their objectivity and rigor. Conveners’ such anticipatory work that transcended the process-outcome tension encouraged collaborators to understand the planning frameworks and their shortcomings, and to share their inputs accordingly. When asked about the conveners’ analyses, most collaborators shared critical comments such as: “They [conveners] use an advanced analytic tool, but they are still tools, right? And they use approximations of reality and there are limits to what they can do with that” (Dan, director of a policy advocacy group). Understanding potential limitations of regional planning allowed collaborators to think about how their own perspectives could fill the gap, whether it was for the process, outputs, or other aspects of the planning process. Consequently, as Randy (a member of an environmental nonprofit) said, IOC meetings often generated vastly different ideas and suggestions: “It’s a lot of work to figure out how to triangulate the different inputs from different groups... it’s incredibly time-consuming and difficult, and we go through an iterative and non-ending cycle of communication.” Promoting tension 3: Creating a third space through boundary work. My analysis showed that the conveners’ management of this tension began as a reactive both-and strategy in the early stage of the IOC, specifically through vacillating between informal and formal engagements to accommodate different collaborators’ needs. They valued both forms of engagements and blurred the boundary between their space as conveners and as data analysts, based on the belief that their work in one place helped to advance their work in the other. But, they realized that separate formal-informal meetings were insufficient to promote productive dialogues and caused misunderstanding among collaborators. So, about a half way through the IOC cycle, they decided to create a third space (Putnam et al., 2016) to be more proactive in encouraging dynamic collaborator interactions and synergy. To illustrate, in the early phase of the IOC, the conveners embraced opportunities to engage with collaborators informally, in addition to their formal meetings. Terry recalled when leaders of an

26 advocacy organization spontaneously came to discuss the validity of some planning technology. Even though Terry was busy and sensed that the collaborators did not have sufficient background knowledge, he still proceeded with the meeting because he believed that “there shouldn’t be a high barrier of entry for someone to engage with us.” Terry provided them in-depth explanations about the technology, which he could not have offered in formal meetings due to time constraints. He saw the informal meeting as an opportunity to hear the collaborators’ thoughts more closely. But, the conveners, including Terry, began to realize the need for better venues to exchange such insights, because those who could not be present at both types of meetings (due to schedule conflicts or staff shortages) would inevitably miss the information and opportunity to build on one another’s ideas. Through several brainstorming sessions, the conveners organized two new formats to encourage more inclusive and dynamic engagements among collaborators. The first was an open house-style meeting, where conveners stood by various “information booths” that allowed collaborators to share ideas and ask questions that they would be hesitant to bring up in formal IOC meetings. The open house also had a forum where collaborators could gather and discuss any topics freely. Tiffany said, “People were like, ‘oh, that’s not [emphasis added] what I’ve been hearing all these months, so I am glad we are having this chance to talk now.’” The second was a retreat-style workshop. Collaborators met at a park and participated in a series of activities designed to facilitate discussions on some challenging issues. Mandy, a participant in the workshop, described it as “a very contentious day”; but, activities such as an icebreaker game that asked collaborators to share stories about “what their ideal region looks like” helped them realize how they “all come from the same place of wanting to help grow nice places for people to live.” Despite some heated discussions, collaborators left the workshop with “a feeling of camaraderie.” These new forms of engagement were characterized as neither formal nor informal; but, they allowed dynamic exchanges of ideas and emergence of tensions in hybrid settings, offering benefits of

27 both informal-formal meetings and opportunities to achieve more than what each form could allow. Organizing such discursive space was enabled by conveners’ desire to connect the boundaries for where they performed as moderators, data analysts, “listening ears,” and any other roles needed to enact open communication processes, instead of neatly classifying their professional world. Tiffany explained one drawback of the new forms of engagement: decision-making happened much more slowly compared to formal IOC meetings, as the conveners often “had to turn on a dime” to meet deadlines following the workshops. Nevertheless, the conveners valued the rich discussions and energy from the workshops, and reported their plan to organize them again in the future. Comparison of the Two IOCs as Tensile Structures and Their Collaborative Capacity Through either-or practices that were aimed at preventing tensions, conveners in Region A organized the IOC such that who participates, what to discuss, and how to engage was clear and did not fluctuate very much. When tensions occurred unexpectedly (e.g., collaborators shared opposing views), the conveners considered them “roadblocks.” As the coding results show (Table 1), in Region A, (a) the sectoral/functional diversity of collaborators was lower; (b) the level of consensus was higher; (c) convener readiness was lower than Region B. Although the causal direction between conveners’ tension management and the coded variables cannot be confirmed based on my data, the descriptive coding results support the findings discussed thus far. The relatively lower level of collaborative capacity does not (and was not intended to) indicate that Region A’s IOC failed; but, based on the factors considered for the purpose of this study, they were less able to generate diverse inputs and dynamic engagements from the IOC compared to Region B. In terms of their IOC as a tensile structure, I would argue that the loss of tension poles (e.g., the lack of participation from advocacy groups, discussions on the process, and informal engagements) reduced the space that collaborators could have used to explore and engage with tensions (see the shapes shown in the third column of Figure 2).

28 By contrast, Region B conveners organized their IOC through more-than practices to promote higher levels of complexities and dynamic changes. When tensions caused practical challenges such as delays, the conveners did not see them as surprises or disruptions. Instead, as examples showed earlier (e.g., Katie and John’s responses to adversary groups and unexpected requests), they remained calm— somewhat nonchalant—and encouraged dialogues to generate alternative ideas. Reflecting on the findings, I would argue that the conveners’ proactive more-than approaches (i.e., not defining what constitutes good collaborators; encouraging collaborators to discuss various aspects of planning; and creating a third space for more dialogues) contributed to “pulling” of the tension poles into different directions to create a tensile structure as well as keeping the discursive space open and dynamic, maximizing the IOC’s collaborative capacity (see the shapes in the second column of Figure 2). Discussion This study contrasted two IOCs and demonstrated what it looks like when organizers try to prevent versus promote tensions as they set up and manage collaborative processes. The findings showed that preventing tensions by removing potential sources of tensions can contribute to making the IOC process neat and easily manageable, while promoting tensions can bring practical challenges due to unexpected, divergent ideas that could alter existing processes. However, preventing tensions can shape the IOC to resist emergent opportunities, whereas promoting tensions allows room for exploring different possibilities. As noted earlier, this study was not aimed at evaluating which of the two IOCs achieved greater success in their regional planning, as it is not feasible to compare how well the two geographically and politically different regions will develop in the next 25 years according to their plans. The goal was to compare how the two IOCs differed in their capacities to gather diverse inputs—a key purpose of IOCs in regional planning—in relation to the conveners’ organizing practices. The findings of this study make three contributions to the study of IOCs and organizational tensions.

29 First, this study extends the organizational tension literature by theorizing tension management as proactive practices from the perspective of organizers. As argued in the introduction, past research has mainly theorized about individuals’ tension management as reactive strategies and overlooked whether and how organizers might intentionally set up for tensions to (or not to) emerge. This study demonstrated that conveners’ proactive management of organizing tensions contributed to preventing or promoting tensions in IOCs, which made a different in both how much tensions emerged during IOCs and how well collaborators were able to engage with the emergent tensions. The proactive perspective would not reject the notion that tensions are normal and unavoidable in any organizational settings (Trethewey & Ashcraft, 2004); but, it would suggest that some situations involve more tensions than others, in part, due to how organizers set the stage for tensions through their practices. By taking the proactive perspective, this study offered insights into what motivates people to take different approaches to tensions. While the rich literature on tension management has identified a variety of ways in which people deal with tensions (see Putnam et al., 2016), it has not been clear why people choose the different strategies. This study found that conveners’ proactive tension management was motivated by their desire to organize their own work in certain ways through their routine practices— identity work (Watson, 2008), anticipatory work (Barley, 2015), and boundary work (Kreiner et al., 2009). This suggests that individuals’ proactive approaches to tensions are informed by their anticipated needs and challenges in their future communicative environment. This finding highlights human agency for influencing how tensions affect their experiences, compared to the reactive view that emphasizes people’s responses to tensions that have already emerged inevitably. Although this study focused on conveners’ proactive tension management, examining both organizers’ proactive and participants’ reactive strategies will help to tease out tensions that are enabled by organizational structures and those that emerge from interactions, and their interplay. We may reconsider Lewis et al.’s study findings (2010), for example, which revealed that IOCs involve structural

30 as well as relational tensions. Their study showed how collaborators dealt with the two types of tensions separately, but it could consider how conveners might have enabled the relational tensions through the way they proactively managed the structural tensions. For example, Region B conveners’ intentional inclusion of groups with divergent ideas likely gave rise to a relational tension (e.g., me- versus weorientations). Researchers can then examine how collaborators work through the tensions reactively and whether their interactions generate new, unintended tensions. Taking this integrative approach to examining the interplay between proactive and reactive strategies will reveal “how different sets [of tensions] function at the forefront or background of organizing” (Putnam, 2015, p. 20). The second contribution of this study is reconceptualization of IOCs as tensile structures. The tensile structure metaphor offers a novel framework for theorizing the mechanism through which tensions support collaboration, specifically when and how. The key finding of this study that promoting tensions—instead of resolving them—can stimulate collaborative interactions, is consistent with findings of past research (e.g., Gibbs, 2009; Koschmann & Laster, 2011). Yet, scholars have also indicated the importance of resolving differences in collaborators’ expertise and meanings (see Bechky, 2003; Carlile, 2004), making it unclear when and how tensions can be advantageous. The tensile structure metaphor suggests that tensions can contribute to IOCs when the used “material,” or the focal issue, has high tensile strength and allows for “stretching” it into different directions. The context of the present study, regional planning, is a good example of such a wicked problem that does not have clear causes and/or solutions; due to its high tensile strength, Region B conveners’ promotion of tensions helped to generate dynamic engagements and to reimagine regional plans from various perspectives. But, if this study was conducted in IOCs with well-defined outcomes, such as implementing an existing plan and making decisions on funding distribution, promoting tensions (e.g., inviting those who oppose the plan or the use of funding) would have rather caused distractions from achieving the specified outcomes. That is, not all IOCs require tensions but those that aim to address open-ended problems.

31 The tensile structure framework also helps to theorize how tensions can contribute to IOCs’ collaborative capacity based on conveners’ organizing practices. Figuratively speaking, Region A conveners’ either-or practices (choosing one of two tension poles) made the tensile structure (IOC) lose its shape and reduce the amount of space for collaborators’ engagements with tensions. Both-and approaches (i.e., selecting both poles, but one at a time or trying to maintain a perfect balance between the two) can increase the discursive space of IOCs, but they are unreliable for maximizing collaborative capacity as they can cause temporary collapses of the tensile structure. As shown in Region B’s case, conveners managed the formal-informal engagements tension through a both-and approach initially, but they recognized the opportunities they missed when they vacillated between the two options. More-than practices used by Region B conveners (reframing, transcending, and creating a third space) can be compared to actively pulling both tension poles into opposite directions to create a tensile structure. Importantly, simply pulling the tension poles is not sufficient to be a more-than practice as it will not differ from “balancing” through a both-and approach. Conveners, while pulling both tension poles to construct a tensile structure (i.e., equally valuing competing options), ought to make efforts to move beyond what the tension poles offer and cultivate new possibilities—like building managers having to maintain and renovate the built space to ensure its usability. Merely selecting both poles without such efforts can lead to stagnation, where diverse ideas cannot generate transformative solutions or move the IOC toward its collaborative goals. To ensure that IOCs’ collaborative capacity is maximized and also utilized in productive ways, conveners need to reflect on their practices and create opportunities to keep the discursive space dynamic, despite some unintended consequences (e.g., collaborators’ questions that they are not prepared to answer). For these reasons, more-than practices require time and skills that can be difficult for organizational members (Putnam et al., 2016). Although I discussed the three types of tensions separately in the findings for conceptual clarity, the tensile structure metaphor suggests that they are inevitably interrelated. As shown in Figure 2, the

32 loss of one tension pole affects how much room is left for the other tensions to play out, ultimately changing the whole shape of the tensile structure. To illustrate, when conveners in Region A selected objective experts over biased experts, having the collaborators who shared “objective” (data-driven) views on regional planning led their discussion to focus on the outcomes over the process, due to shared belief that statistical modeling outputs help to best predict the future; in turn, conveners felt that formal engagements were sufficient for their IOC. Considering how the tension poles are located in different corners but are connected to create a common space for an IOC, how one set of tension poles is set up will influence and be influenced by how other poles function. The third contribution of this study is practical implications for conveners’ mindful organizing. Since organizing IOCs as tensile structures is useful for addressing complex issues that require diverse ideas, conveners should first evaluate the main goal of their IOC in terms of its “tensile strength” (e.g., if it allows room for ideas to clash until the goal is achieved or if reaching consensus is a priority). If the desired result is to tackle a “wicked problem” (high tensile strength), conveners should resist the urge to simplify their assumptions about effective IOC processes or to control situations to meet their expectations when organizing. Instead, they should prepare for contingencies and develop resilience to deal with unexpected events and expected challenges (Weick & Suttcliffe, 2006). To promote tensions successfully, it is important for them to treat their work as negotiated and adaptive, rather than definite and immutable. Engaging in reflective questions such as “What were we doing when we made the decision?” “Should we do more or less of what we did?” and “Was our anticipation spot-on or partially blind?” (Forester, 2012; Minei & Bisel, 2013) can assist their mindful organizing. The tensile structure metaphor can serve as a conceptual guide for conveners to visualize what kind of space they are creating through setting up the tension “poles” (competing options) as they make structural decisions. Limitations

33 This study is not without limitations. Since my field research began during the last year of the four-year IOC cycle in regional planning, I used retrospective interviews to understand how conveners set up the IOC initially. Examining the entire cycle of collaboration could have allowed for a dynamic analysis of how the tensions emerged, changed, and were managed differently at different points along the process. Another potential limitation is that my cross-case comparative analysis prevented me from assessing how either-or approaches used by conveners in Region A may have contributed to some goals unique to the region. Even though Region A generated less diverse inputs compared to Region B, perhaps either-or practices were appropriate given the region’s unique local contexts. Future studies can examine whether, how, and what types of contextual factors (e.g., strong “wind” affecting tensile structures, such as political climate) make a difference in conveners’ abilities to promote tensions, even when the IOCs are organized to address a wicked problem (a material with high tensile strength).

34 References Apker, J., Propp, K. M., & Zabava Ford, W. S. (2005). Negotiating status and identity tensions in healthcare team interactions: An exploration of nurse role dialectics. Journal of Applied Communication Research, 33(2), 93-115. doi:10.1080/00909880500044620 Barley, W. C. (2015). Anticipatory work: How the need to represent knowledge across boundaries shapes work practices within them. Organization Science, 26(6), 1612-1628. doi:10.1287/orsc.2015.1012 Baxter, L. A., & Montgomery, B. M. (1996). Relating: Dialogues & dialectics. New York, NY: Giuldford Press. Bechky, B. A. (2003). Sharing meaning across occupational communities: The transformation of understanding on a production floor. Organization Science, 14(3), 312-330. doi:10.1287/orsc.14.3.312.15162 Canales, R. (2013). Weaving straw into gold: Managing organizational tensions between standardization and flexibility in microfinance. Organization Science, 25(1), 1-28. doi:10.1287/orsc.2013.0831 Carlile, P. R. (2004). Transferring, translating, and transforming: An integrative framework for managing knowledge across boundaries. Organization Science, 15(5), 555-568. doi:10.1287/orsc.1040.0094 Christensen, L. T., Morsing, M., & Thyssen, O. (2015). Discursive closure and discursive openings in sustainability. Management Communication Quarterly, 29(1), 135-144. doi:10.1177/0893318914563574 Cummings, J. N. (2004). Work groups, structural diversity, and knowledge sharing in a global organization. Management Science, 50(3), 352-364. doi:10.1287/mnsc.1030.0134

35 Dean, M., & Oetzel, J. G. (2014). Physicians’ perspectives of managing tensions around dimensions of effective communication in the emergency department. Health Communication, 29(3), 257-266. doi:10.1080/10410236.2012.743869 Doerfel, M. L., & Taylor, M. (2004). Network dynamics of interorganizational cooperation: The Croatian civil society movement. Communication Monographs, 71(4), 373-394. doi:10.1080/0363452042000307470 Dorado, S., & Vaz, P. (2003). Conveners as champions of collaboration in the public sector: A case from South Africa. Public Administration and Development, 23(2), 141-150. doi:10.1002/pad.270 Driskill, G. W., Meyer, J., & Mirivel, J. C. (2012). Managing dialectics to make a difference: Tension management in a community-building organization. Communication Studies, 63(2), 243-261. doi:10.1080/10510974.2011.634476 Fairhurst, G. T., Cooren, F., & Cahill, D. J. (2002). Discursiveness, contradiction, and unintended consequences in successive downsizings. Management Communication Quarterly, 15(4), 501540. doi:10.1177/0893318902154001 Forester, J. (2012). On the theory and practice of critical pragmatism: Deliberative practice and creative negotiations. Planning Theory, 12(1), 5-22. doi:10.1177/1473095212448750 Gibbs, J. (2009). Dialectics in a global software team: Negotiating tensions across time, space, and culture. Human Relations, 62(6), 905-935. doi:10.1177/0018726709104547 Gray, B. (1989). Collaborating: Finding common ground for multiparty problems. San Francisco, CA: Jossey-Bass. Gray, B., & Purdy, J. (2014). Conflict in cross-sector partnerships. In M. M. Seitanidi & A. Crane (Eds.), Social partnerships and responsible business: A research handbook (pp. 205-225). Abingdon, UK: Routledge.

36 Jian, G. (2007). "Omega is a four-letter word": Toward a tension-centered model of resistance to information and communication technologies. Communication Monographs, 74(4), 517-540. doi:10.1080/03637750701716602 Koschmann, M., & Laster, N. M. (2011). Communicative tensions of community organizing: The case of a local neighborhood association. Western Journal of Communication, 75(1), 28-51. doi:10.1080/10570314.2010.536965 Kreiner, G. E., Hollensbe, E. C., & Sheep, M. L. (2009). Balancing borders and bridges: Negotiating the work-home interface via boundary work tactics. Academy of Management Journal, 52(4), 704730. doi:10.5465/AMJ.2009.43669916 Lewis, L., Isbell, M. G., & Koschmann, M. (2010). Collaborative tensions: Practitioners' experiences of interorganizational relationships. Communication Monographs, 77(4), 460-479. doi:10.1080/03637751.2010.523605 Lüscher, L. S., & Lewis, M. W. (2008). Organizational change and managerial sensemaking: Working through paradox. Academy of Management Journal, 51(2), 221-240. doi:10.5465/AMJ.2008.31767217 Medved, C., Morrison, K., Dearing, J., Larson, S., Cline, G., & Brummans, B. (2001). Tensions in community health improvement initiatives: Communication and collaboration in a managed care environment. Journal of Applied Communication Research, 29(2), 137-152. doi:10.1080/00909880128107 Minei, E., & Bisel, R. (2013). Negotiating the meaning of team expertise: A firefighter team’s epistemic denial. Small Group Research, 44(1), 7-32. doi:10.1177/1046496412467830 O'Reilly, C. A., & Tushman, M. L. (2013). Organizational ambidexterity: Past, present, and future. The Academy of Management Perspectives, 27(4), 324-338. doi:10.5465/amp.2013.0025

37 Pepper, G. L., & Larson, G. S. (2006). Cultural identity tensions in a post-acquisition organization. Journal of Applied Communication Research, 34(1), 49-71. doi:10.1080/00909880500420267 Purdy, J. M. (2012). A framework for assessing power in collaborative governance processes. Public Administration Review, 72(3), 409–417. doi:10.1111/j.1540-6210.2011.02525.x Putnam, L. L. (2015, July). Moving beyond ‘both-and’ approaches: Alternative strategies for managing paradoxical tensions. In 31st Colloquium of the European Group on Organization Studies, Athens, Greece. Putnam, L. L., Fairhurst, G. T., & Banghart, S. (2016). Contradictions, dialectics, and paradoxes in organizations: A constitutive approach. Academy of Management Annals, 10(1), 65-171. doi:10.1080/19416520.2016.1162421 Smith, W. K., & Lewis, M. W. (2011). Toward a theory of paradox: A dynamic equilibrium model of organizing. Academy of Management Review, 36(2), 381-403. doi:10.5465/amr.2009.0223 Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory (2nd ed.). Thousand Oaks, CA: Sage Publications. Tomlinson, F. (2005). Idealistic and pragmatic versions of the discourse of partnership. Organization Studies, 26(8), 1169-1188. doi:10.1177/0170840605055338 Trethewey, A., & Ashcraft, K. L. (2004). Special issue introduction: Practicing disorganization: The development of applied perspectives on living with tension. Journal of Applied Communication Research, 32(2), 81-88. doi:10.1080/0090988042000210007 Tracy, S. J. (2004). Dialectic, contradiction, or double bind? Analyzing and theorizing employee reactions to organizational tension. Journal of Applied Communication Research, 32(2), 119146. doi:10.1080/0090988042000210025 Watson, T. J. (2008). Managing identity: Identity work, personal predicaments and structural circumstances. Organization, 15(1), 121-143. doi:10.1177/1350508407084488

38 Weber, E. P., & Khademian, A. M. (2008). Wicked problems, knowledge challenges, and collaborative capacity builders in network settings. Public Administration Review, 68(2), 334–349. doi:10.1111/j.1540-6210.2007.00866.x Weick, K. E., & Sutcliffe, K. M. (2006). Mindfulness and the quality of organizational attention. Organization Science, 17(4), 514-524. doi:10.1287/orsc.1060.0196 Woo, D., Putnam, L. L., & Riforgiate, S. E. (2017). Identity work and tensions in organizational internships: A comparative analysis. Western Journal of Communication, 81(5), 560-581. doi:10.1080/10570314.2017.1312510 Wood, D. J., & Gray, B. (1991). Toward a comprehensive theory of collaboration. The Journal of Applied Behavioral Science, 27(2), 139-162. doi:10.1177/0021886391272001 Yin, R. K. (2014). Case study research: Design and methods (5th ed.). Thousand Oaks, CA: Sage.

39 Table 1 Examining Collaborative Capacity of IOCs in Region A and Region B: Coding Procedures and Outcomes Criteria

Definition and Rationale

Collaborator Definition: The extent to which collaborators Diversity differed from conveners in terms of their organizational sector and functional expertise. Rationale: The more structurally and functionally diverse the collaborators are, the more the value of input-sharing increases (Cummings, 2004); conveners who tried to promote tensions were more likely to involve collaborators with different backgrounds, increasing the IOC’s collaborative capacity.

Region A 17.6%

Region B 48%

(b) Functional diversity: The percentage of collaborators who had different expertise (i.e., not trained in regional planning or related fields)

29.4%

32%

Operationalization and Coding Procedure (a) Structural diversity: The percentage of collaborators who worked in different organizational sectors (i.e., not in government sector)

Level of Consensus

Definition: The degree to which collaborators and conveners shared unanimous beliefs about a given issue. Rationale: The lower the level of consensus, the more conveners encouraged collaborators to offer diverse perspectives and challenge their ideas (i.e., promoting tensions).

The percentage of collaborators who expressed the same views as conveners on the newly adopted collaboration tool. Collaborators’ responses to the question, “What do you think about this tool?” were categorized in terms of how similar their views were to their conveners’ views, using one through three scale (1 = different, 2 = neutral, 3 = same as conveners’). Then, I calculated the percentage of the collaborators who were coded as three (i.e., those who had the same views).

41.6%

24.2%

Convener Readiness

Definition: The extent to which conveners were prepared and willing to engage directly with collaborators. Rationale: The higher convener readiness is, the more conveners expected and encouraged dynamic collaborative processes (through the way they enacted their role), contributing to higher collaborative capacity of the IOC.

(a) Conveners’ relational history: The percentage of conveners who directly engaged with collaborators. Each convener was categorized based on his/her reported experience interacting with collaborators (1 = yes, have worked with them directly, or 2 = no, have not). Then, I calculated the percentage of the conveners who were coded as one. (b) Conveners’ attitude: The percentage of conveners who expressed positive attitude toward engaging with collaborators. Each convener’s interview data, specifically how they described their feelings toward engaging with collaborators, was categorized by using one through three scale (1 = negative, such as reluctance or hesitation, 2 = neutral expressions, 3 = positive attitude such as willingness, enthusiasm, or acceptance that it is an important part of their routine work). Then, I calculated the percentage of conveners who were coded as three.

57.1%

66.7%

30.8%

73.3%

40 Figure 1 Illustration of Tensile Structure

Figure 2 A Loss of a Tension Pole and the Resulting Shape of Tensile Structure (Reduced Space) IOC with two major types of tensions (A-A1, B-B1)

When all four tension poles are pulled into different directions

B

When one tension pole is lost (e.g., selection of A1 over A) A

A1

1 A

A1

1

B B1

B1

IOC with three major types of tensions (A-A1, B-B1, C-C1) B

C

A

1 B1

When one tension pole is lost (e.g., selection of A over A1) A1

A

A1

C1

When all six tension poles are pulled into different directions

B

C C1

B1